University of Massachusetts Boston
ScholarWorks at UMass Boston
Graduate Masters Theses
Doctoral Dissertations and Masters Theses
8-1-2012
Implications of Self-Other Overlap in Unsuccessful
Romantic Relationships
Josephine A. Bell
University of Massachusetts Boston
Follow this and additional works at: http://scholarworks.umb.edu/masters_theses
Part of the Clinical Psychology Commons
This Open Access Thesis is brought to you for free and open access by the Doctoral Dissertations and Masters Theses at ScholarWorks at UMass
Boston. It has been accepted for inclusion in Graduate Masters Theses by an authorized administrator of ScholarWorks at UMass Boston. For more
information, please contact library.uasc@umb.edu.
Recommended Citation
Bell, Josephine A., “Implications of Self-Other Overlap in Unsuccessful Romantic Relationships” (2012). Graduate Masters Theses.
Paper 119.
IMPLICATIONS OF SELF-OTHER OVERLAP IN UNSUCCESSFUL ROMANTIC
RELATIONSHIPS
A Thesis Presented
by
J. ANNA BELL
Submitted to the Office of Graduate Studies,
University of Massachusetts Boston,
in partial fulfillment of the requirements for the degree
MASTER OF SCIENCE
August 2012
Clinical Psychology Program
© 2012 by J. Anna Bell
All rights reserved
IMPLICATIONS OF SELF-OTHER OVERLAP IN UNSUCCESSFUL ROMANTIC
RELATIONSHIPS
A Master’s Thesis Presented
by
JOSEPHINE A. BELL
Approved as to style and content by:
________________________________________________
Sheree D. Conrad, Senior Lecturer
Chairperson of Committee
________________________________________________
Jean Rhodes, Professor
Member
________________________________________________
Alice S. Carter, Program Director
Member
_________________________________________
Alice S. Carter, Program Director
Clinical Psychology Program
_________________________________________
Carol Smith, Chairperson
Psychology Department
iv
ABSTRACT
IMPLICATIONS OF SELF-OTHER OVERLAP IN UNSUCCESSFUL ROMANTIC
RELATIONSHIPS
August 2012
J. Anna Bell, B. A., University of Oregon
M.S., University of Oregon
Directed by Senior Lecturer Sheree D. Conrad
The social psychological concept of self-other overlap describes how identities
and cognitive representations of people have a tendency to merge as they become closer
to one another. Because greater self-other overlap tends to be associated with such
positive characteristics as closeness and intimacy, it has generally been considered a
desirable trait in relationships. In a previous study (Bell, 2009), preliminary evidence
supported the idea that there may in fact be negative consequences to having higher
levels of self-other overlap in relationships with domestic violence, including diminished
self-esteem and life satisfaction. The current study expands on Bell (2009) study and
examines the implications of greater self-other overlap within three different samples:
students currently in a romantic relationship, students that recently ended a romantic
relationship, and women from a domestic violence shelter who recently ended a violent
relationship.
v
TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION……………………………………………………………..
1
Background Study………………………………………………………
13
Overview of Current Study…………………………………………..
15
II. METHOD ……………………………………………………………………….
19
Participants ……………………………………………………………….
19
Procedure………………………………………………………………….
20
Measures…………………………………………………………………..
22
III. RESULTS………………………………………………………………………
26
IV. DISCUSSION…………………………………………………………………
31
Conclusion ………………………………………………………………..
36
REFERENCES……………………………………………………………………………….
38
1
CHAPTER I:
INTRODUCTION
People are generally motivated by a strong desire to establish and maintain
relationships with others. According to the need to belong hypothesis, creating
relationships is not only something that people seek out and pursue but is also a
fundamental and pervasive need (Baumeister & Leary, 1995). According to this
hypothesis individuals seek to establish positive relationships with others that are
reciprocal and supportive in nature. Success in establishing these connections can have
serious implications in terms of the amount of support received. For example, high levels
of social support have been associated with several positive characteristics such as
increased levels of happiness and well-being and can provide a buffer between stressful
events and negative mental health outcomes (Baumeister, 1991; Baumeister & Leary,
1995; Cohen & Wills, 1985; Sarason, Sarason, & Gurung, 1997). Conversely, the
absence of this support can often have negative effects, as the lack of support has been
associated with higher levels of stress, anxiety, and depression (Cohen, Sherrod, & Clark,
1986).
The strength of the support received is directly related to the quality (and not the
quantity) of social connections. Simply having a series of casual relationships with
different individuals is not sufficient to fulfill this innate need to belong, since superficial
interactions do not carry with them the same positive effects that would otherwise result
from enduring, quality relationships (Cohen et al., 1986). Studies have demonstrated that
2
having individuals upon whom one can rely and feel supported by can serve as effective
buffers, whereas having a broad social network alone is not sufficient (Cohen &
Hoberman, 1983).
Not only do we seek out connections to other people, but once these relationships
are created we are extremely averse to losing them. In some cases individuals may go to
great lengths to maintain these bonds, even when remaining in these relationships runs
contrary to our needs, desires, and even safety (Baumeister & Leary, 1995; Strube &
Barbour, 1983). Such situations may be as benign as remaining with a partner who does
not meet or appreciate one’s needs, but individuals are often motivated to maintain
relationships even when in an abusive relationship that leaves them exposed to violence
and harm. Reluctance shown by many when ending relationships and the tendency of
some individuals to stay with an abusive partner are illustrations of how strong our
motivations can be in preserving our relationships. Although it can be difficult to
understand from the perspective of someone outside the relationship, the benefits
obtained from maintaining such relationships may, in some cases, outweigh the costs.
Although people establish relationships due to their desire for meaningful
connections with others, it is theorized that they are also motivated to form these
connections due to an underlying desire to add to who they are as individuals. Aron and
Aron (1986) have developed a theoretical framework known as the “self-expansion
model” that seeks to explain some of the motives behind relationship formation.
According to this theory, this expansion occurs in two parts: expansion and integration.
Through expansion, people want to increase complexity within their sense of self by
3
adding novel experiences. This acquiring new information in the expansion phase is
often followed by a desire to reduce the complexity of this experience by incorporating
this new information into the already established self-concept, a process known as
integration. It is through both of these parts that one’s sense of self grows and develops.
If people seek out relationships in an attempt to expand their self-concept,
acquiring connections with others can be emotionally rewarding because it fulfills this
motivation. In fact, there is evidence to support this idea, as the formation of new
relationships, both platonic and romantic, tends to be experienced as pleasurable. Indeed,
self-expansion seems to occur rapidly in the beginning stages of relationships, since it is
during this time that there is a great exchange of information (Strong & Aron, 2006).
Additionally, it is during the process of establishing relationships that the lines between
self and other can be blurred and elements or characteristics of one person may be
adopted or influenced by the other.
When people expand their sense of self by entering into a relationship they may
adopt and integrate the resources, perspectives, and characteristics of their partner in the
relationship and include them in their self-concept, increasing the degree to which the self
and the other person “overlap” (Aron, Aron, Tudor, & Nelson, 1991). For example, if one
relationship partner is considered to be more artistic, through activities such as engaging
in art-related discussions or attendance to art-related functions, the other, previously less-
artistic partner might come to consider his or herself more artistic over the course of the
relationship. By including their partner in their sense of self, individuals are, in a sense,
blurring the boundaries between themselves and their partners as part of a process known
4
as self-other overlap. The existence of self-other overlap within romantic relationships
can be especially pronounced, as the exchange of information, self-disclosure, and time
spent together tends to be very high (Aron, Mashek, & Aron, 2004; Strong & Aron,
2006).
In an attempt to measure the amount of the self that overlaps with a partner, Aron,
Aron, and Smollan (1992) developed a single-item measurement known as the Inclusion
of Others in the Self (IOS) Scale. Using this scale, individuals were asked to rate overlap
the amount of overlap they perceived themselves as sharing with a relationship partner.
Participants were provided with seven increasingly overlapping circles (one circle labeled
“self” and the other labeled “other”), and asked to indicate which of the figures best
represented their perceived relationship with a partner. Based on findings from this
measure, higher levels of self-other overlap were shown to be associated with a variety of
positive characteristics. Greater overlap on this measure has been correlated with greater
closeness, intimacy, commitment, and relationship quality (Aron et al., 1992; Mashek,
Aron, & Boncimino, 2003). The extent to which people indicated having self-other
overlap with their romantic partners was also related to relationship robustness; self-other
overlap was predictive of whether relationships were still intact three months later (Aron
et al., 1992).
Other researchers have constructed alternative ways of measuring the overlap that
occurs in close relationships. While Aron and colleagues have developed a way to
measure self-other overlap using a visual depiction, Davis, Conklin, Smith, and Luce
(1996) have developed a measure that assesses cognitive depictions of this overlap. By
5
utilizing adjective lists, Davis et al. (1996) asked participants to select which adjectives
described themselves and which described another target. The number of adjectives that
participants marked on both lists (essentially the number of characteristics that individual
shares with their partner) indicates amount of perceived overlap that is shared between
the two partners (Davis et al., 1991). This research was motivated by the desire to
understand how perspective-taking might influence the amount of self-other overlap that
one perceives as having with another. Within these experiments it was demonstrated that
perspective-taking does increase the amount of self-other overlap, as measured by the
percentage of overlapping adjectives.
The measure constructed by Davis et al. (1991) provides a different way of
measuring self-other overlap that involves characteristics that participants see themselves
as sharing with their partner. Although the adjective checklist that was developed by
Davis et al. (1991) is a measure that has traditionally been used to evaluate the level of
self-other overlap that follows perspective-taking, the current study uses it in the realm of
close relationships by investigating its connection to self-esteem. The reasoning behind
this usage is that within the literature both the IOS scale and the adjective checklist have
generally been viewed as measuring the same concept, because both of these
measurements are believed to be accessing self-other overlap. If these measurements are
in fact measuring similar things as the literature has implied, then utilizing the adjective
checklist in close relationships, a domain in which it has not been typically used (but is a
domain in which the IOS if often used), should yield similar results. If the IOS and
adjective checklists end up being related to different things, then it stands to reason that
6
these different measurements may produce different results when it comes to assessing
different elements of self-other overlap.
There is increasing evidence to support the idea that these different measures of
self-other overlap may actually be measuring somewhat different things. Myers and
Hodges (in press) found evidence that self-other overlap might fall along two different
dimensions that they termed “perceived closeness” and “overlapping representations.”
The amount of perceived closeness was represented by the extent to which individuals
feel close to the other person. Overlapping representations can be represented by the
number of traits and characteristics that individuals perceives themselves sharing with
another person. Myers and Hodges also found that perspective taking influenced one of
these dimensions (“perceived closeness”) more than the other (“overlapping
representations”), supporting the idea that these different measures of self-other overlap
may actually be measuring separate constucts that occur within relationships.
Although people are strongly motivated to seek out and maintain relationships
with others (and romantic relationships in particular), it is inevitable that some
relationships will not be successful. Due to the fact that self-expansion is both generally
sought-after and rewarding, the loss of those connections when a relationship ends can be
upsetting. Furthermore, within the domain of self-other overlap the loss of that close and
meaningful connection does not necessarily mean that those elements of the former
partner that were incorporated into the self-concept are quickly and cleanly eliminated. It
may be the case that elements of the former partner remain a part of the self-concept of
7
the person left behind long after the relationship has been terminated, which may make it
more difficult for some individuals to “move on” after that relationship has ended.
Exploring the dissolution of romantic relationships, Agnew and Etcheverry (2006)
examined how much individuals often still felt connected to their former relationships
and relationship partners. The researchers conducted a longitudinal study in which three
groups of individuals were assessed in regards to their commitment level and cognitive
interdependence: those who stayed in their relationship, those who left their relationship,
and those who were left by their partner. Within this study, cognitive interdependence
was considered the extent to which participants felt they lost some of their sense of self
when the relationship ended. Not surprisingly, those who terminated the relationship
became less committed to their former partner and showed less cognitive
interdependence with their former relationship and partner over time. Interestingly, those
who were left by their partner showed no different pattern in their levels of commitment
to or interdependence with their former relationship and partner than those who were
currently in a relationship. The implication of this result is that some individuals who
experience the end of their relationship appear to cognitively consider their relationship
with their former partners as still intact, at least for a while.
Although the partners that are left behind may have some self-concept confusion
directly following the dissolution of a relationship, this experience is not permanent.
Eventually, individuals are forced to face the loss of their partner and that connection,
and are left to assess which aspects of themselves are their own and which are shared
with their former partner. Slotter, Gardner, and Finkel (2009) argue that individuals
8
experience a change in their self-concept in the aftermath of a breakup and may have
difficulty identifying which aspects of themselves are known enduring parts of their sense
of self (a concept identified as “self-concept clarity”) and which are elements of their
former partner as they seek to regain a stable self-concept. Slotter et al. (2009) found that
not only were individuals’ self-concepts less clear following a break-up, their self-
concepts were also experienced as being smaller (self-concept constriction), and the
content and perception of those selves was also changed. In some cases constriction and
clarification of the self-concept can take time, since it tends to occur only when
individuals have begun to accept the reality of the breakup and take steps to move on
from that former relationship (Lewandowski, Aron, Bassis, & Kunak, 2006). Because
there are aspects of the self that an individual might have shared with a former partner,
goals, social circles and other activities may be changed or lost due to the dissolution of
that relationship. As a direct result of those changes, individuals may see themselves as a
person without some of the identities and opportunities that they had within the
relationship, resulting in a reduced self-concept.
Additionally, the loss of close attachments is commonly associated with
depression and guilt among those left by their partner, as individuals might blame
themselves for the loss of the relationship (Jack, 1999). For those who did not initiate the
break-up, coming to terms with rejection can also prove to be difficult to overcome
(Sbarra, 2006; Slotter et al., 2009). These factors undoubtedly have implications for both
self-esteem and well-being, as both of these can be negatively impacted in the aftermath
of a break-up. It is possible that individuals may begin to see themselves as not being
9
good enough for the love and support that was previously offered by the partner, or may
hold oneself responsible for the partner leaving.
Although the aftermath of a relationship can be difficult in that individuals have
to come to terms with the stress of losing a relationship, there may be relationships that
are in themselves extremely stressful, as is the case when there is the presence of extreme
conflict and/or abuse. While relationships in general tend to be complicated and
confusing, with many factors contributing to their formation and maintenance,
relationships characterized by domestic violence often seemed especially perplexing. To
an outside observer it may seem illogical that any individual would stay in a relationship
that leaves her susceptible to physical harm. Those experiencing domestic violence are
not generally kept behind barred windows or locked doors and so it may be difficult to
comprehend why these victims would find it so hard to leave (Herman, 1992). However,
the ties that bind the battered to the batterer are no less real to the victim.
Research on women in heterosexual relationships1 indicates that the victim
encounters many daunting obstacles if she is to leave an abusive partner. One of the most
common reasons why a woman may stay with an abusive partner is lack of economic
resources outside the relationship. In fact, past research has shown that economic
dependence in the abusive relationship is correlated with the desire of women to stay with
their abusive partner (Strube & Barbour, 1983). With a controlling and violent partner, a
woman who leaves an abusive environment may find herself with no money, no car, and
1 Although there are abusive relationships that involve women abusing men and
sometimes occurs within same-sex relationships, the focus here is on the most prevalent
of abusive relationships: male perpetrators and female victims in heterosexual
relationships (Straus & Gelles, 1986).
10
no place to go (Lundberg-Love & Wilderson, 2006). Although there may be shelters for
her to take refuge in, she may not be aware of their existence and there is no guarantee
that there will be space for her once she gets there. Additionally, the shelters that are
available generally only offer a spot for a limited amount of time, after which the woman
must move on. These women may also have (a very legitimate) fear of a violent outburst
from their partners in retaliation for their leaving, given the level of violence already
present within the relationship. This fear is fully supported by the literature; often the
most dangerous point in an abusive relationship is when the woman decides to leave
(Oths & Robertson, 2007). Furthermore, a woman with children may also want to stay
with her abusive partner because she wants to keep their family intact, fearing what may
happen if she tried to survive on her own or wanting to keep a father figure in her
children’s life.
Psychologically, there are a number of reasons a woman might stay in an abusive
relationship. A victim might feel compelled to remain in the relationship out of the
affection and love she feels for her partner. Although on the surface this might not seem
to be an adequate excuse for staying in a violent relationship, it is important to note that
many relationships with domestic violence begin like most others: two individuals are
brought together through attraction and a relationship is established through courtship
and is built on mutual affection, attraction, and love (Gelles & Straus, 1988; Herman,
1992; Walker, 1979). In this way, these violent relationships, at least in the beginning
stages, are very similar (if not indistinguishable) from those that do not contain violence.
In fact, only a small percentage of victims in abusive relationships actually report
11
violence at its beginning stages (Walker, 1979). By the time the relationship is
established, it may be too late for the victim to easily or safely leave the perpetrator.
Furthermore, a victim may not feel that her abuser is at fault for the violence.
Research has shown that women who were living with abusive partners are very likely to
blame themselves for the violence they have been subjected to (Andrews & Brewin,
1990; Walker, 1979). In fact, the tendency of victims to defend even identify with the
very people that cause them harm is a well-documented phenomenon. For example,
Stockholm Syndrome, or “hostage syndrome”, was originally developed to describe
individuals who identified with their captors in hostage situations and is a process that
includes rationalizing the violence or minimizing the abusive experience by trying to see
things from the perspective of the perpetrator (Graham, Rawlings, & Rimini, 1988;
Lundberg-Love & Wilderson, 2006). This generally results in hostages or victims
finding fault with their own actions and thus excusing and legitimizing the perpetrator’s
violent behavior.
Although one may argue that Stockholm Syndrome may be the result of an
abnormal, unique, and stressful situation, these same underlying forces may also be at
work in a domestic setting. Even within family settings the same processes might be at
work. After all, it is not uncommon for children to defend and even protect their abusive
parents (Green, 1996). Like captives in a hostage situation, children who are dependent
on the very person that is abusing them are stuck in a situation from which there is no
escape. Women in situations of domestic violence can also find themselves in a similar
position. Some of the women in domestic violence situations are, by chance or design,
12
economically dependent upon their abusive partner. The fewer financial resources a
woman has in a relationship, the more dependent she is on her partner, and the more
power her partner has (Anderson, 1997; Walker, 1979). Similarly, in an attempt to gain
power, many batterers use fear to dominate their partners and destroy their autonomy
(Anderson, 1997; Herman, 1992). In this way, women in domestic violence situations
tend to resemble victims in hostage situations (Graham et al., 1988).
Even if victims of domestic violence have the capacity to physically escape their
abusive partner, the alternatives may be unfavorable or even non-existent. Perhaps a
woman’s relationship with her abusive partner has truly destroyed other relationships
with friends, eliminating possible social support. There is the possibility that her
relationship with her partner resulted in her losing her job, or perhaps she was never
employed, thus leaving her with few personal resources and little financial security
beyond the support that she receives from her abusive partner.
If a victim resorts to identifying with her perpetrator, it could result in anticipating
the abuse and possibly taking steps to lessen (but not necessarily avoid) the abuse or its
impact. At the same time, these actions might cause victims to see the abusive situation
from the perpetrator’s perspective, reducing and even justifying the partner’s violent
behavior (Walker, 1979). Excusing or rationalizing the abuse may hinder any
improvements in the relationship and may make the victim less willing to leave the
abusive situation. Being able to take the perspective of her abusive partner may be
beneficial temporarily, but may be harmful in the long run if a woman is unable to
separate herself from her partner emotionally and physically. Notably, this sharing of
13
perspectives with a partner is the second element mentioned in the self-expansion model
and is one of the foundations of self-other overlap (Aron et al., 1991).
The self-expansion model (Aron & Aron, 1986) may also be valuable in
explaining how violent relationships begin, because these relationships resemble non-
violent ones in the earlier stages (Walker, 1979). However, the positive relationship
characteristics associated with self-other overlap found in healthy relationships could
potentially break down within an abusive environment. One does not necessarily want a
close, intimate, and enduring relationship with someone who is physically abusive, and
staying with an abusive partner may have negative consequences for the victim’s physical
safety in addition to well-being, resulting in lower levels of self-esteem (Lystad, Rice, &
Kaplan, 1996).
Background Study
The majority of studies addressing the topic of self-other overlap have been done
using college students; other potentially problematic relationships have not been
explored. Simply being enrolled and actively taking classes in a university setting is not
a luxury that all individuals enjoy, particularly those with unreliable incomes and
sporadic living situations. In a study that was part of an undergraduate thesis, the role of
self-other overlap in relationships was explored recruiting a sample of women in a
domestic violence shelter who had recently ended their relationships with their abusive
partners and were actively seeking services and support (Bell, 2009). Due to the fact that
the majority of the research on self-other overlap in relationships has been conducted on
relatively healthy relationships, Bell’s (2009) study is unique in that it explores how
14
overlap, which is generally considered a benefit, may actually have negative implications
when present in the aftermath of a violent and abusive relationship.
In line with research conducted by Myers and Hodges (in press), Bell’s (2009)
study conducted on formerly abused women revealed that the two measures of overlap
(the IOS scale and the adjective checklist) were correlated with different outcomes,
providing more evidence to support the idea that different measures of overlap might not
be measuring the same thing. Within this sample, Bell found that the IOS scale was
significantly correlated with abused women’s intention to return to their former abusive
partners (as measured by the Intent to Return Questionnaire; Stanley & Markman, 1992);
the more self-other overlap as measured by the IOS, the greater the intent to have further
interaction with the former partner.
For the other measures of self-other overlap, items from the adjective checklist
were divided by valence (positive or negative), and the number of adjectives that the
abused woman endorsed for both herself and her partner was computed. Percentages of
positive or negative adjectives were the result of the number of positive (or negative)
adjectives that a woman endorsed for both herself and her partner, divided by the total
number of positive (or negative) adjectives a woman endorsed for herself. Using this
method of computation, Bell found that women who had a higher percentage of positive
adjectives in common with their former partners had lower levels of self-esteem.
Conversely, those who indicated having a higher percentage of negative adjectives in
common with their former partners indicated having higher levels of self-esteem.
15
Meanwhile, the alternative measure of self-other overlap (the IOS scale) was not
correlated with self-esteem.
It was hypothesized that the latter results might be due to the fact that these
women “lost” part of themselves (those characteristics) when they left their partner.2
Those who considered themselves as sharing many positive characteristics with their
partner may have seen themselves as leaving those qualities behind with their partners or,
alternatively, may currently perceive those positive qualities as somehow being “tainted”
or tied up with their abusive partners. Meanwhile, those who saw themselves as sharing
many negative characteristics with their partner may have felt that they may have left
behind those traits or moved beyond those negative qualities, thereby resulting in
increased self-esteem. However, it is important to note that the participants did not
appear to have completely “lost” these traits, because they did still label these traits as
descriptive of themselves at the time of the study.
Overview of Current Study
Due to the difficulty of accessing a sample of domestic violence victims, Bell’s
initial research was conducted on a small sample (N=21). The current study was able to
add additional subjects to the earlier Bell sample, providing the opportunity to see if a
larger sample would support previously observed effect sizes. Additionally, because no
previous work addressed how self-esteem might be influenced by self-other overlap as
measured by Davis et al.’s adjective checklist, it remains unclear which aspects of Bell’s
2 Because all the participants were seeking services from a domestic violence shelter at
the time of participating in the study, it can be safely assumed that all of these women left
their partner and were actively taking steps to distance themselves from that abusive
environment.
16
(2009) findings were unique to women who had just left abusive relationships. As a
result, these findings could be driven by the experience of abuse, or might be the result of
breaking up with a romantic partner. It also might be the case that these results are found
in all relationships, including ones that are ongoing. In the conclusion of that study, it
was clear that additional research with non-abused women as a comparison group was
needed to investigate those possibilities and the current study was designed to explore
these unanswered research questions.
In this study we seek to investigate the relationship between self-other overlap,
self-esteem, and well-being. Additionally, we will be exploring whether there are
particular relationship characteristics are driving the results; that is to say, whether the
presence of abuse and relationship status can change the relationship between these
variables. Three groups of participants were studied in an attempt to understand which
relationship characteristics are driving these results. The first group was an expansion of
the sample collected by Bell (2009) as part of her undergraduate thesis and was
composed of a small number of women who recently left an abusive relationship and
were currently seeking and/or utilizing services in a domestic violence shelter. We
hypothesized that when a greater percentage of positive adjectives are shared between
subjects and the former partner, participants will have lower levels of self-esteem and
well-being and that in contrast, those with more negative adjectives in common with their
former partners will show the opposite effect with higher levels of self-esteem and well-
being. Due to the fact that the abused women are in a shelter suggesting that the
separation is quite recent, they may still be actively working on establishing an identity
17
separate from that of their partner and may not yet be mentally separated from their
former partners, it is predicted that these women still have much of their self-concept
associated with their former partner. For the same reason, we also hypothesize that the
formerly abused women who indicate having more self-other overlap as measured by the
IOS will have higher depression scores and lower self-esteem and life satisfaction. The
presence of abuse in these relationships may contribute to these detrimental effects
because very important aspects of the relationship and the women’s role in it may come
to be defined by the abusive partner.
The second group consists of college students who have recently experienced a
break up with a romantic partner. Similar to the sample of abused women, this group of
participants is also expected to have low levels of self-esteem and higher depression
scores with increased self-other overlap on the IOS, but to a lesser degree than the
formerly abused women because the effect of abuse on mental health outcomes is likely
to be absent, although participants were not selected on the basis of abuse history. We
anticipate these negative effects will result due to the fact that breaking up with someone
is often an unpleasant experience in general. Also, we anticipate this sample to have
results similar to the domestic violence sample on the adjective checklist: We believe that
self esteem might be enhanced due to shared negative adjectives, while self esteem might
be lessened due to a sharing of positive adjectives. At the same time, due to the
assumption that these relationships are not abusive in nature, we do not expect these
results to be as strong as they would be in the sample of formerly abused women.
18
Our final sample consisted of college students currently in a romantic
relationship. Within this group we predict results that are similar to the previous work
that has been done on self-other overlap. That is to say, those who have higher levels of
self-other overlap (as measured by the IOS) will have increased self-esteem and life
satisfaction. Contrary to our other samples of individuals who are no longer in a
romantic relationship, we anticipate that a high percentage of positive adjectives in
common with their current partner will be associated with higher levels of self-esteem,
while a high percentage of negative adjectives in common will be associated with lower
levels of self-esteem. The reasoning for this hypothesis is that individuals who believe
that they share a number of positive characteristics with their current partner might feel
better about both themselves and their relationship.
19
CHAPTER II:
METHOD
Participants
Participants were drawn from two different populations; one group consisted of
31 participants who were women recruited from a women’s shelter in Eugene, Oregon
and the second consisted of 327 undergraduates who attended the University of Oregon.
The majority of women who were recruited from the women’s shelter (21 participants)
were collected as part of an undergraduate thesis (Bell, 2009). For the purposes of this
project, data from 10 additional women from that women’s shelter were added. The
average age of the participants from the shelter was 34 (SD =10.74), with ages ranging
from 18 to 62. Twenty-one of the women listed having at least some college, while nine
listed having a high school diploma (or equivalent). One participant indicated never
attending high school. Twenty-five participants were Caucasian, three identified as
Native American and three identified as Hispanic.
Of the 327 student participants drawn from the university, 238 were female and
89 were male. All of the students participated in this study for partial fulfillment of
course credit for classes in either the psychology or linguistics department. Using
prescreening measures, participants were divided into one of two conditions depending
